---
license: cc-by-4.0
datasets:
- zirui3/webMedQA-instructions
- zirui3/med-image-reports
---

# Medical Vision-Language Models

# Model
  * LLM: LLaMA
  * Visual Encoder: BLIP2

# dataset
* Medical Text
  * PubMedQA
  * MedQA
  * MedDialog
  * [webMedQA](https://huggingface.co/datasets/zirui3/webMedQA-instructions)
* Medical Image-Text Pairs
  * XRay reports
    * MMIC-CXR 370k
    * openi 60k